Run a report's data part and keep what its ARD holds

Runs the study's setup code, the report's ARD code and its normalization and rework (data_lines()) in a fresh R process, from the study folder, and reads ard_meta() from the result: the normalized data when the rework ran, the raw ard otherwise. A report whose ARD its ARD definition makes reads its rows of the study ARD when they are made from the definition as it is now (ard_status(): built); else its ARD is made here, in a folder of its own, and the study ARD (output/ard/ard.rds) is not written – making the study ARD is update_study_ard()'s, and the official run's. The data frames the code loaded (adsl, ...) add what the ARD lacks: each variable's label, and the order of its values (a factor's levels, or the order of a numeric companion <name>N: TRT01A by TRT01AN). The result is kept in tflplanner's home, so it is there the next time the study is opened.

Usage

fetch_ard(study, output_id, timeout = 300, home = tflplanner_home())

ard_info(study, output_id, home = tflplanner_home())

ard_data(study, output_id, home = tflplanner_home())

ard_raw(study, output_id, rows = FALSE, home = tflplanner_home())

ard_preview_state(
  study,
  output_id,
  meta = ard_info(study, output_id, home),
  home = tflplanner_home()
)

Arguments

study: An rtfstudy.
output_id: The report.
timeout: Seconds to allow.
home: tflplanner's home. ard_info() returns the metadata kept, ard_data() the normalized data kept with it (what the builder's preview is planned from), ard_raw() the ARD before its normalization (rows = TRUE: the report's rows as the study ARD has them, with their ids), and ard_preview_state() whether what is kept is still the report's ARD: "missing", "outdated" (its definition, its code, the study's setup or the data have changed since) or "current".
rows: TRUE: the report's rows with their ids, as the study ARD has them (the ARD definition's reports; NULL for the others).
meta: What was kept (ard_info()); by default, read.

Value

The metadata (ard_meta()) with fetched (the time), source ("data" or "ard"), error (the rework's error, if it failed), made ("here": made for the report; "study": read from the study ARD), print (what it was made from, ard_preview_state()) and log, invisibly.